[Coco] MPI 2D rendering

[Gene Heskett]

On Saturday 28 March 2015 16:34:43 RETRO Innovations wrote:
> On 3/28/2015 1:50 PM, Gene Heskett wrote:
> > On Saturday 28 March 2015 14:39:03 RETRO Innovations wrote:
> >> For those who care, I've sent the MPI design off for prototype
> >> manufacture.  Here is a view of the unit in more natural coloring:
> >>
> >> http://gerblook.org/pcb/5j7wfp8V8PJ5pt6BH8kNzQ
> >>
> >>    * Final boards will be blue or black, as I'm not a fan of green,
> >> but green is cheapest for a prototype.
> >>    * Astute viewer will notice the larger board with the sockets is
> >>      "upside" down.  This is by design, as a stencil will need to
> >> be created, and the front of the buffer board and the back of the
> >> main board have the majority of the SMT items.  When the boards are
> >> separated, it won't make any difference.
> >>
> >> Jim
> >
> > Looks good Jim.  Do patterns for a box exist yet?
>
> Well, no, but here are the dimensions:
>
>   * Overall (x*Y) =  3.85x6"
>   * Assuming bottom left corner of board is 0,0, 10 .187" holes
> centered @
> o one column @ .275" 
> o another @ 3.575" (dimension - .275", board pattern is
> symmetric in X dimension)
>       o 5 rows @ 1.1", 2.25", 3.4", 4.55, 5.7"  (rows every 1.15")
>   * As you can see from the pics, the standoffs define the size of the
>     cart rectangle (1.1 + (.187/2) is the front of the first cart, and
>     2.25"-(.187/2) defines the bottom of it.)
>   * The carts are wider than the PCB, though, so plan accordingly for
>     the width of the enclosure

A diskcon cart is 4.245" wide and .915" thick.  What I had in mind for a 
first try was a 3 layer sandwich, a layer with cart sized cutouts  about 
1/4" above the socket, another nominally 1.35" above that, also with 
cart cutouts, and another w/o the cart cutouts on the bottom.  
Clear .090" thick lexan because I can get that at lowes (but at their 
price) cutting all mount holes in one stack, and about centered on a  
5.5" wide panel.  Get the connectors on the sides measured & all lined 
up & cut a front and back side, and a left & right end, solvent welded 
into a box.  Sides & ends nominallly 2.5" wide, left end maybe two 
pieces so one can be installed after the cable slot is milled out, 
possibly screwsed into a corner bracket so it can be dissed again if the 
cable needs changed. 

Stack that up on an assembled board and adjust the standoffs till its 
about 2.5" thick.
>       o Note that those are minimums for cart size.  Measure a cart to
>         be sure, add some space, then subtract from the dimensions
> above to get "fit" amount.  Divide by 2 and subtract from dimensions
> above to get cut lines.
>   * All connectors are at front (Y = 0) side:
>       o DIN5 is centered at X=.35", Z=.375" above top of PCB (PCB is
>         1.6mm in thickness).  hole diameter needs to be at least .55",
>         but I would not recommend that, as the DIN plug thickness will
>         be > 0.  CoCo uses .666" diameter holes
>       o Mini USB is centered at X=3.3", with opening minimum of
> X=.303"* Z=155" with bottom of connector resting on top of PCB
> directly (1.6mm from top of standoffs).  I would highly recommend
> adding .01 to .02 to both dimensions for fit.
>       o Coax is centered at X=3.65, but I do not have diameter or Z
> axis data on it yet.  I can find the datasheet if needed.
>       o 80 conductor micro IDC cable needs to have a slot in the case
>         about .3" above the top of the PCB plane in the Z axis,
> centered on the board (X=3.85/2), for the width of 2 IDE cables
> stacked on top of each other.
>
> That should give you enough to create a mechanical drawing. If someone
> wants to create mechanicals, I've been itching to try out a light
> gauge steel enclosure option, and this might be a good candidate. 
> Obviously, all other things assumed working, this would require a few
> folks to consider it before we ran a batch.

I like that idea as the steel could be grounded, then serving as some of 
the shielding to meet FCC specs? I don't have a copy of a current 47 CFR 
to see what the current part 15 or part 19 might have to say in that 
regard.  But with the signal traces as narrow as they are, surrounded by 
a ground plane, I'd have to think it would be far less of an rf 
radiation field to measure than the huge traces used in the OEM versions 
with a vanishly small amount of ground plane in comparison that they had 
were.

Not that I'm going to fuss, I mention it just so its not a surprise if it 
gets brought up.  I don't normally look for that sort of trouble. With 
the switch to digital doing away with all but one tv station in the 
country (we're it) in the low band (2-6) vhf, I certainly do not see a 
nationwide problem with it.  We're it, because at the time they were 
assigning digital channels, their computer wasn't programmed to protect 
the National Radio Observatory at Green Bank WV.  When they originally 
gave us channel 58, I called the interference agent at Green Bank & 
threw it at him, asking how much noise we could make on channel 58.  I 
had to translate that to megahertz for him, but the limit of power we 
could send his way before he would shut us down was 4.78 Watts!

I asked for a confirming letter, then took that to the Vegas NAB show a 
few weeks later and went around to all the transmitter makers asking for 
quotes on a channel 58 transmitter with 4.78 Watts rms output.  Watching 
their reactions when they finally groked I was serious 
was "entertaining". We had some jollies and drank a stock tank full of 
beer in the Townsend booth after floor hours telling war stories amongst 
us engineers, then I handed our lawyer several copies of it & our owner, 
Russ told him to "fix" it.  So we stayed on channel 5, and channel 12 
stayed in their old channel too, a low enough frequency we didn't bother 
Green Bank.  But now we have a new problem as an off air broadcaster.  
No one is now making a low band vhf consumer tv antenna!  And we're 
bleeding eyeballs profusely as the wind destroys the old ones that have 
been up for almost 50 years now.

So as Paul Harvey said, now you know the rest of the story, if anybody 
cares...

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>